The Hybrid Security Framework to Mitigate DDoS and Other Attacks in Vehicular Ad-Hoc Networks (VANETs): Vulnerability Assessment, Model Design as Well as Performance Evaluation -A Review

Vehicular Ad-hoc Networks (VANETs) are also the building blocks of intelligent transportation systems because they allow safety-critical applications such as real-time communication between vehicles and roadside infrastructure. Nevertheless, VANET environment is very susceptible to cyber-attacks like Distributed Denial of Service (DDoS), Sybil, replay and routing attacks because it is decentralized and has an open wireless communications media. The review will provide an in-depth discussion of the algorithmic paradigms of new hybrid security frameworks that combine machine learning-based intrusion detection, trust management schemes, and cryptographic authentication strategies to improve the security of VANET. The recent studies (2022-2026) comparative analysis has shown that the intrusion detection systems are characterized by a detection accuracy of 85-90, whereas machine learning-based methods have the capability to detect intrusion with an average of 93-96. The hybrid security models that incorporate both trust evaluation, encryption and deep learning models have shown the highest performance in terms of detection rates of 97.8-99.2, low rates of false positive of 2-4, and high rates of packets delivery of 88 percent to 96 percent when exposed to DDoS attack conditions. What is more, the hybrid model minimizes end-to-end delay by about 18 percent relative to traditional security means. The findings prove the hypothesis that multi-layer hybrid security architecture can offer scalable, precise, and real-time security to next-generation VANET and Internet of Vehicles (IoV) environments.